Rethinking treatment for non-metastatic deficient mismatch repair colorectal and small bowel cancers

Abstract

BACKGROUND

Deficient mismatch repair (dMMR), present in approximately 15% of colorectal cancers (CRC), defines a tumor subset with sensitivity to immune checkpoint inhibitors (ICIs). Landmark clinical trials have established ICI therapy as a transformative treatment for non-metastatic dMMR CRC; however, evidence from routine clinical practice remains limited.

AIM

To describe real-world clinical, radiological, and pathological responses (pCR) to ICI therapy in patients with non-metastatic dMMR colorectal and small bowel adenocarcinoma (ADC), outside clinical trial settings.

METHODS

This retrospective, single-center observational study included eight consecutive patients with non-metastatic dMMR colorectal or small bowel ADC treated with ICI therapy between January 2023 and June 2026. Mismatch repair (MMR) status was determined by immunohistochemistry for MLH1, MSH2, MSH6, and PMS2. Treatment response was assessed using computed tomography, magnetic resonance imaging and/or positron emission tomography/computed tomography, endoscopy, and digital rectal examination, as appropriate according to tumor. Complete clinical response (cCR) was defined as the simultaneous absence of radiological, endoscopic, and clinical evidence of disease. Complete pCR was defined as ypT0N0 after surgery. All treatment decisions were made by a multidisciplinary tumor board.

RESULTS

Eight patients (four females, four males; median age 59.5 years) were included: Four colon cancers, three locally advanced rectal cancers (LARC), and one small bowel ADC. Two patients had known lynch syndrome. ICI regimens included pembrolizumab, dostarlimab, nivolumab, and ipilimumab. Among patients with LARC, two achieved a sustained cCR and remained on watch-and-wait for more than 20 months; the third underwent surgery for stenosis and achieved pCR. The three patients with locally advanced colon cancer underwent minimally invasive surgery and achieved pCR. One frail patient with colon cancer achieved cCR and is managed non-operatively. The patient with small bowel ADC remained radiologically disease-free at 12 months. No grade ≥ 3 immune-related adverse events necessitating treatment discontinuation were recorded.

CONCLUSION

ICI therapy achieves high rates of complete response in non-metastatic dMMR colorectal and small bowel ADC treated in routine clinical practice, facilitating organ-preservation and minimally invasive surgery. Systematic upfront MMR testing at diagnosis is essential to avoid treatment delays and ensure timely identification of candidates for these treatments.

Key Words: Colorectal cancer; Mismatch repair deficiency; Immune checkpoint inhibitors; Rectal cancer; Small bowel neoplasm; Pembrolizumab; Dostarlimab; Nivolumab; Lynch syndrome

Core Tip: This real-world retrospective study describes eight patients with non-metastatic mismatch repair (MMR)-deficient colorectal and small bowel adenocarcinoma treated with immunotherapy in different clinical settings. Favorable responses were observed across all tumor subsites. Complete clinical responses enabled a watch-and-wait strategy in some patients with locally advanced rectal cancer. In colon cancer immunotherapy facilitated minimally invasive surgery with pathological complete responses in all operated patients. Delayed MMR testing postponed treatment initiation in some cases, highlighting the need for systematic upfront immunohistochemical assessment at diagnosis. These findings support the feasibility of implementing immunotherapy-based strategies in routine surgical oncology practice beyond clinical trial settings.

INTRODUCTION

Colorectal cancer (CRC) ranks third in global cancer incidence and second in cancer-related mortality[1,2]. In Spain, it is the most frequently diagnosed malignancy, with approximately 15000 annual deaths[2]. Small bowel adenocarcinoma (ADC), on the other hand, is a rare malignancy; its rarity limits the available evidence regarding its optimal treatment and the number of disease-specific clinical trials, of which very few have been conducted[3].

The advent of molecular tumor profiling has brought about a fundamental shift in cancer treatment towards precision medicine. The expression of key proteins can determine a patient’s response to different treatment modalities[4].

Immunohistochemical (IHC) assessment of the mismatch repair (MMR) system-through analysis of the four proteins MLH1, MSH2, MSH6, and PMS2-is a key determinant of therapeutic strategy, since the loss of the MMR protein deficient MMR (dMMR) status confers sensitivity to immune checkpoint inhibitors (ICIs). Approximately 15% of CRC harbor dMMR[4,5], a finding that carries distinct prognostic and predictive implications depending on the disease stage. Therefore, CRC is not a single entity, but rather a heterogeneous group of diseases with different molecular profiles, prognoses, and responses to treatment.

In metastatic CRC, ICIs have demonstrated clear superiority over conventional chemotherapy in dMMR selected patients[6-9]. Building on this success, ICI-based strategies were extended to the neoadjuvant setting in localized dMMR CRC.

Clinical evidence for neoadjuvant immunotherapy initially emerged in patients with locally advanced rectal cancer (LARC). Cercek et al[10] reported a 100% complete clinical response (cCR) rate with dostarlimab monotherapy in dMMR patients, with sustained responses and no surgery required at the time of publication. Updated results with a larger cohort confirmed durable organ preservation[11]. These findings have made neoadjuvant ICI therapy the preferred option in LARC dMMR according to current NCCN guidelines[12] and have been confirmed in multicenter series[13,14].

The evidence has evolved from rectal cancer to colon cancer. For locally advanced colon cancer, the NICHE[15] and NICHE-2[16] trials have demonstrated remarkable rates of pathological response (pCR) with dual immune checkpoint inhibition, consolidating neoadjuvant immunotherapy as an emerging strategy in this field.

Therefore, in recent years, conventional management in certain settings involving dMMR CRC and small intestine cancer has been changing following the results of high-impact studies. Importantly, no consensus has yet been reached regarding the optimal ICI regimen, number of cycles, or post-treatment protocol.

Randomized clinical trial results, however, are derived from carefully selected populations that may not reflect the heterogeneity encountered in everyday practice. Real-world series are therefore essential to evaluate feasibility, characterize response patterns across clinical scenarios, and assess outcomes under routine conditions.

Here, we present a real-world experience of ICI therapy across three intestinal subsites-colon, rectum, and small bowel-in patients with non-metastatic dMMR disease.

MATERIALS AND METHODS

Study design and patients

This is a retrospective observational study of consecutive patients with non-metastatic dMMR colorectal, or small bowel ADC treated with ICI at a single center between January 2023 and June 2026.

The study was conducted in accordance with the Declaration of Helsinki and the European General Data Protection Regulation.

Ethics committee approval was obtained with a waiver of informed consent, given the retrospective design and use of anonymized clinical data. Reporting follows the STROBE guidelines.

Inclusion criteria: Histologically confirmed colorectal or small bowel ADC; non-metastatic disease (TNM staging, AJCC 8^th edition); dMMR confirmed by IHC-obtained either from the index lesion or, when a biopsy of the index lesion was not available, from tumor tissue of prior oncological resections in patients with known Lynch syndrom; At least one ICI cycle administered with neoadjuvant intent; and age ≥ 18 years.

Exclusion criteria: Metastatic disease (stage IV) at diagnosis; and MMR status unavailable by IHC.

Variables

Demographic variables (age, sex), clinical variables (tumor location, Lynch syndrome history, comorbidities, presenting symptoms), and diagnostic variables (radiological, endoscopic, and histological findings including IHC results) were recorded. Treatment variables included ICI agent, dose, number of cycles, and immune-related adverse events. The choice of ICI regimen (monotherapy vs combination) and number of cycles was individualized at multidisciplinary tumor board (MTB).

Response assessment was performed using computed tomography, magnetic resonance imaging, and/or positron emission tomography/computed tomography, endoscopy (colonoscopy or rectoscopy), and, for rectal tumors, digital rectal examination.

Response definitions: CCR in rectal cancer required simultaneous complete radiological, endoscopic, and clinical response, with negative biopsy where performed[10]. In colon cancer, cCR required complete endoscopic response with negative biopsy and complete radiological response. In small bowel cancer, cCR was based on imaging alone. pCR was defined as the absence of viable tumor cells in the primary tumor bed and regional lymph nodes (ypT0N0)[15,16].

Statistical analysis

Given the descriptive nature of this case series, only descriptive statistics were performed. Continuous variables are expressed as median and interquartile range or full range. Categorical variables are expressed as absolute n and (%).

RESULTS

During the study period, eight patients received ICI: Four with colon cancer, three with LARC, and one with small bowel ADC. A patient flowchart is shown in Figure 1. All treatment decisions were made by an MTB including specialists in General and Digestive Surgery, Medical Oncology, Radiation Oncology, Radiology, Nuclear Medicine, and Pathology.

Figure 1  Patient flowchart.

Demographic and clinical characteristics

The series included four female and four male patients, with a median age of 59.5 years (range 33-85). Two patients (P1, P6) had known lynch syndrome at the time of diagnosis. Detailed demographic and clinical data are provided in Table 1.

Table 1 Baseline demographic and clinical characteristics of the included patients.

Patient	Tumor location	Age (years)	Sex	Medical history	Known lynch	Symptoms
P1	Small bowel	85	Male	Colon cancer (surgery 1990/2015), prostate cancer (2015)	Yex (MSH2)	Asymptomatic
P2	Colon	78	Male	COPD, OSA, moderate pulmonary hypertension, atrial fibrillation, IC-FEr (FE 35%), Moderate exertional dyspnea	No	Asymptomatic
P3	Colon	71	Female	No	No	Constitutional symptoms, malnutrition, and abdominal mass
P4	Colon	77	Female	No	No	Constitutional symptoms, anemia, malnutrition, and abdominal mass
P5	Colon	33	Male	No	No	Anemia
P6	Rectum	36	Female	First-degree lynch syndrome (MSH6)	Yes (MSH6)	Progressive weight loss and a change in bowel habits
P7	Rectum	37	Male	Grade III astrocytoma (surgery 2008/2010)	No	Rectal bleeding and a change in bowel habits
P8	Rectum	48	Female	No	No	Rectal bleeding and a change in bowel habits

COPD: Chronic obstructive pulmonary disease; OSA: Severe obstructive sleep apnea; IC-FEr (FE 35%): Heart failure with reduced ejection fraction (left ventricular ejection fraction 35%).

Diagnostic findings

Radiological, endoscopic, and histological findings at diagnosis are summarized in Table 2.

Table 2 Baseline imaging, endoscopic, and histopathological findings.

	CT/MRI	PET/CT	Colonoscopy	Biopsy	IHC
P1	Concentric mural thickening with ulceration and locoregional involvement	Increased metabolic activity	-	-	-
P2	No histopathological findings	Focal hypermetabolic deposit near the hepatic flexure of the colon	Flat elevated lesion with central ulceration, irregular morphology, and a disorganized glandular pattern	ADC	dMMR phenotype (loss of MLH1 and PMS2 expression)
P3	Locally advanced colon cancer, with a necrotic lymph node mass adjacent to the ileocolic region and infiltration into the abdominal wall musculature	-	A mass at the hepatic flexure with extensive, deep ulceration	Poorly differentiated ADC	dMMR phenotype (loss of MLH1 and PMS2 expression) BRAF V600E mutation
P4	Circumferential wall thickening involving the cecum and ascending colon up to the hepatic flexure, with extension into the pericolic fat and ileocolic lymphadenopathy	-	Extensive ulcerated mass from the cecum to the hepatic flexure, involving the ileocecal valve and approximately 75% of the circumference	Poorly differentiated ADC	dMMR (loss of PMS2 expression) BRAF V600E mutation
P5	Wall thickening of the cecum with ileocolic lymphadenopathy	-	Nodular mass at the ileocecal valve involving approximately 75% of the circumference	Moderately differentiated ADC	dMMR (loss of MSH2 and MSH6 expression)
P6	Tumor located 9 cm from the anal verge, staged as cT3bN1b, with EMVI and a threatened CRM (< 1 mm)	Rectal mural thickening with regional lymphadenopathy and no other areas of uptake	Ulcerated lesion extending from 7 to 12 cm from the anal verge	ADC	dMMR (loss MSH6 expression)
P7	Lower rectal cancer staged as cT3b cN1a	-	Ulcerated, eroded tumor located 3-4 cm from the anal verge, involving approximately 50% of the circumference without stenosis	ADC	After CRT and first consolidation CT: DMMR phenotype (loss of PMS2 expression)
P8	Tumor staged as T3bcN2b, located 6 cm from the anal verge, with EMVI positive, threatened CRM < 2 mm, and no lateral lymph nodes	Rectal tumor with no other findings	Stenosing rectal tumor located 9 cm from the anal verge	ADC	After CRT: DMMR (MSH2-, MSH6-)

EMVI: Extramural vascular invasion; CRM: Circumferential resection margin; CT: Computed tomography; MRI: Magnetic resonance imaging; PET/CT: Positron emission tomography/computed tomography; IHC: Immunohistochemical; ADC: Adenocarcinoma; CRT: Chemoradiotherapy; dMMR: Deficient mismatch repair.

Notably, patient 1 (small bowel ADC) had no biopsy of the index lesion; dMMR status had, however, been confirmed by IHC on the resection specimens of his prior colorectal surgeries (loss of MSH2 expression, consistent with his known Lynch syndrome), and the MTB decision to initiate ICI therapy was based on this previously documented dMMR phenotype together with the patient's surgical and oncological history. Among the colon cancer patients, comorbidities and frailty in patient P2 were key determinants of treatment strategy. The remaining three colon cancer patients (3, 4, and 5) presented with locally advanced tumors. MMR status was obtained from the diagnostic biopsy in all colon cancer patients. In contrast, in two LARC patients without known lynch syndrome, MMR results only became available after completion of conventional chemoradiotherapy (CRT), causing a delay in ICI initiation.

Treatment and response

ICI regimens, doses, number of cycles, treatment responses, and subsequent management (surgery or watch-and-wait) are detailed in Table 3.

Table 3 Immune checkpoint inhibitor regimen, immune-related adverse events, treatment response, and subsequent management (surgery or watch-and-wait).

	ICI treatment (agent, dose, number of cycles)	ir-AEs	Treatment response	Subsequent management	AP	Actual situation	Follow-up (months)
P1	Pembrolizumab 2 mg/kg every 21 days, 18 cycles	No	Complete radiological resolution of the mural thickening	W&W	-	Asymptomatic, alive	12 meses
P2	Pembrolizumab 2 mg/kg every 21 days, 10 cycles	Hospitalized for respiratory infection and heart failure exacerbation	Complete radiological and endoscopic response, with no evidence of tumor	W&W	-	Asymptomatic, alive	4 meses
P3	Nivolumab 3 mg/kg and ipilimumab 1 mg/kg every 3 weeks. 4 (N + I) + 4 (N)	No	Radiological improvement with reduction in tumor size	Laparoscopic right hemicolectomy	ypT0N0 (0/28 lymph nodes)	Asymptomatic, alive
P4	Pembrolizumab 2 mg/kg every 21 days, 5 cycles	No	Decreased in size; reduction in the size of the necrotic pathological lymph node adjacent to the tumor	Robotic right hemicolectomy	ypT0N0 (0/60 lymph nodes)	Asymptomatic, alive
P5	Pembrolizumab2 mg/kg every 21 days, 4 cycles	No	Resolution of the wall thickening in the blind section with persistent lymphadenopathy and a slight increase in metabolism	Robotic right hemicolectomy	ypT0N0 (0/30 lymph nodes)	Asymptomatic, alive
P6	Dostarlimab 500 mg every 3 weeks, 9 cycles	No	Complete clinical response on DRE, colonoscopy, CT and PET/CT	W&W	-	Asymptomatic, alive	27 meses
P7	Dostarlimab 500 mg every 3 weeks, 9 cycles (after CRT + XELOX1)	No	Complete clinical response (CCR) on DRE, colonoscopy (scar in the rectum with good endoscopic appearance), and MRI	W&W	-		22 meses
P8	Dostarlimab 500 mg every 3 weeks, 9 cycles (after CRT)	No	Endoscopic stenosis without malignancy	Robot-assisted TME	ypT0N0 (tumor regression 0/4)	Asymptomatic, alive	20 meses

W&W: Watch & wait; TME: Total mesorectal excision; ir-AEs: Immune-related adverse events; CRT: Chemoradiotherapy; ICI: Immune checkpoint inhibitor; CT: Computed tomography; MRI: Magnetic resonance imaging; PET/CT: Positron emission tomography/computed tomography; AP: Anatomical pathology; DRE: Digital rectal examination.

Patient 2 required hospitalization for comorbidity-related complications during treatment but achieved a cCR and was managed non-operatively. Patients 3, 4, and 5, all with locally advanced colon cancer, underwent minimally invasive surgery following completion of ICI therapy. All three achieved pCR (ypT0N0), and none of the patients received adjuvant chemotherapy.

Among the three patients with LARC, two (patients 6 and 7) achieved cCR and have maintained it for more than 20 months under watch-and-wait surveillance. The third patient (patient 8) required surgery because of tumor-related stenosis and underwent robot-assisted total mesorectal excision, achieving a pCR (ypT0N0).

Patient 1, with small bowel ADC, received 18 cycles of ICI therapy. In the absence of a pretreatment biopsy, treatment response could only be assessed radiologically, which contributed to the decision to prolong therapy. The patient remains radiologically disease-free at 12 months of follow-up.

No adverse events requiring treatment discontinuation were recorded in any patient.

Imaging and endoscopic findings

Representative radiological, endoscopic, and histopathological findings at diagnosis and after completion of ICI therapy are shown in Figures 2, 3, 4, 5, 6, 7, 8, and 9. Corresponding to patients 1-8 in Table 1, respectively, each figure integrates all available radiological, endoscopic, and histopathological findings for an individual patient, facilitating a comprehensive assessment of baseline disease and treatment response.

Figure 2 Case 1. A: Axial and sagittal abdominal computed tomography (CT) images obtained at diagnosis; B: Axial and sagittal abdominal CT images obtained after completion of immune checkpoint inhibitor therapy.

Figure 3 Case 2. A: Colonoscopy at diagnosis; B: Colonoscopy after completion of immune checkpoint inhibitor therapy, showing the site of prior tattoo marking.

Figure 4 Case 3. A: Coronal and axial abdominal computed tomography (CT) images at diagnosis; B: Colonoscopy at diagnosis; C: Coronal and axial abdominal CT images after completion of neoadjuvant therapy; D: PET/TAC prior to surgery; E: Intraoperative findings; F: Right hemicolectomy specimen; G: Macroscopic examination of the surgical specimen showing residual lesion.

Figure 5 Case 4. A: Coronal and axial abdominal computed tomography (CT) images and colonoscopy at diagnosis; B: Post-treatment abdominal CT and positron emission tomography/computed tomography after completion of neoadjuvant therapy; C: Right hemicolectomy specimen; D: Microscopy showing an ulcerated cecal tumour bed with fibroinflammatory changes and granulation tissue, without residual viable tumour cells, consistent with a complete pathological response. Reactive lymph node changes and extensive tumour necrosis related to treatment are also observed.

Figure 6 Case 5. A: Coronal and axial abdominal computed tomography (CT) images and colonoscopy at diagnosis; B: Post-treatment abdominal CT, positron emission tomography/computed tomography, and colonoscopy after completion of neoadjuvant therapy; C: Gross pathology of the opened right hemicolectomy specimen showing an excavated lesion at the ileocecal junction extending into the ascending colon, corresponding to the treated tumor bed; detailed view of a 2 cm × 1.5 cm residual lesion at the post-treatment site.

Figure 7 Case 6. A: Colonoscopy, sagittal magnetic resonance imaging (MRI), and axial diffusion-weighted MRI at diagnosis; B: Rectoscopy and MRI at 6 months post-treatment, demonstrating a complete clinical response.

Figure 8 Case 7. A: Colonoscopy, sagittal magnetic resonance imaging (MRI), and diffusion-weighted MRI at diagnosis; B: Rectoscopy showing residual stenosis, and post-treatment MRI demonstrating tumour response.

Figure 9 Case 8. A: Sagittal and axial magnetic resonance imaging (MRI) and colonoscopy at diagnosis; B: Sagittal and axial MRI and colonoscopy after completion of neoadjuvant immune checkpoint inhibitor therapy.

DISCUSSION

The management of CRC and small bowel cancers is evolving rapidly, driven by advances in minimally invasive surgery and by targeted oncological therapies. Molecular tumor profiling has become central to this evolution, enabling personalized treatment selection. In this context, systematic upfront IHC assessment of MMR proteins is no longer optional: It is recommended by current NCCN and ESMO guidelines for all patients diagnosed with colorectal, small bowel, and rectal cancers[12,17-19].

MTBs are essential in this setting. Surgeons must be familiar not only with surgical options, but also with the role of molecular biology in treatment selection, as this proficiency is integral to delivering a truly individualized, coordinated strategy.

ICIs-pembrolizumab, nivolumab, ipilimumab, and dostarlimab-block inhibitory checkpoints that suppress anti-tumor immunity, restoring effective tumor killing[20]. Historically, dMMR status in CRC, which is found in 15% of patients, was associated with a favorable prognosis in early-stage disease but a worse prognosis in advanced disease, a setting in which the benefit from conventional chemotherapy also appears more limited[21]. ICI therapy has transformed this landscape across multiple settings.

In metastatic dMMR CRC, pembrolizumab monotherapy-based on KEYNOTE-177, phase 3 clinical trial[7,9]-and nivolumab plus ipilimumab-based on CheckMate-142 phase 2 clinical trial[6,8]-are established first-line options with durable response rates and manageable toxicity profiles. Thus, pembrolizumab monotherapy is currently established as first-line treatment for patients with metastatic dMMR CRC[19].

Expanding role in non-metastatic disease, neoadjuvant ICI therapy in dMMR LARC is supported by the strongest body of clinical evidence. The 2022 report from Cercek et al[10]-a 100% cCR rate with dostarlimab monotherapy and organ preservation without conventional multimodal treatment-was a landmark finding. Updated data with longer follow-up and larger cohorts have confirmed the durability of responses[11], leading to the adoption of ICI therapy as the preferred neoadjuvant treatment for dMMR LARC in international guidelines[12]. In our series, two LARC patients achieved cCR maintained for more than 20 months. Notably, the delayed MMR result in two LARC patients who had already received CRT illustrates a critical real-world gap. This finding aligns with published multicenter series[13,14] and reinforces current guideline recommendations.

Following the promising results achieved in dMMR LARC, neoadjuvant ICI therapy has also been investigated in patients with locally advanced colon cancer. The NICHE[15] and NICHE-2[16] trials demonstrated pCR rates of 60%-68% with neoadjuvant nivolumab and ipilimumab in dMMR colon cancer, establishing neoadjuvant immunotherapy as an emerging strategy in this setting. In our series, all three operated patients with locally advanced colon cancer achieved pCR, regardless of the specific ICI regimen administered-consistent with the lack of regimen consensus reported in recent reviews[4,5]. Phase 3 trials are ongoing to define the optimal neoadjuvant protocol[22]. The success of organ-preserving strategies in dMMR LARC raises the question of whether selected patients with dMMR colon cancer may also benefit from non-operative management. In our series, one patient with substantial comorbidity achieved a sustained cCR and was managed without surgery because of his operative risk. Prospective studies are needed to determine whether watch-and-wait can be safely extended to dMMR colon cancer.

Beyond CRC, evidence for ICI therapy in dMMR small bowel ADC remains extremely limited because of the rarity of this tumor, and current management could be largely extrapolated from the experience gained in CRC. Our first patient is notable for receiving neoadjuvant ICI without a biopsy of the index small bowel lesion; the indication was nonetheless supported by a dMMR phenotype previously confirmed by IHC on the resection specimens of his prior colorectal surgeries, in the setting of known lynch syndrome. The exclusive reliance on radiological response assessment, together with the absence of endoscopic or histological endpoints for the index lesion, contributed to treatment extension to 18 cycles. At 12 months of follow-up, the patient remains radiologically disease-free, supporting the rationale for ICI even in this uncommon tumor type. Watch-and-wait strategies remain essentially unexplored in small bowel ADC, where the rarity of the tumor and the absence of prospective data preclude firm recommendations[3]. However, the absence of histological confirmation remains a major caveat that must be weighed carefully in this subgroup.

In our cohort, all treatment decisions were made on an individualized basis at the MTB, and no patient received concurrent chemotherapy plus immunotherapy. In the two patients managed with a non-operative intent (patient 1, small bowel; patient 2, frail colon cancer), treatment was deliberately prolonged, as no consensus exists in the literature regarding watch-and-wait strategies in small bowel or colon cancer[4,5]. In the three patients with locally advanced colon cancer the therapeutic objective was surgery: Patient 3 received dual immune checkpoint blockade based on the CheckMate-8HW trial[23], whereas patients 4 and 5 received pembrolizumab monotherapy based on KEYNOTE-177[7,9]. In patients 7 and 8 the MMR result became available only after completion of CRT, which subsequently led us to systematize MMR determination on the diagnostic biopsy.

From a molecular standpoint, two patients with locally advanced colon cancer (patients 3 and 4) harbored BRAF V600E mutations together with loss of MLH1/PMS2 expression, a pattern characteristic of sporadic dMMR arising through MLH1 promoter hypermethylation rather than of Lynch syndrome[24]. This distinction is relevant for molecular classification and prognostic stratification. In our limited experience, both patients with this sporadic-type profile achieved a pathological complete response after neoadjuvant immunotherapy.

Our study has several limitations. This is a small retrospective case series with inherent selection and reporting biases. Treatment protocols and follow-up schedules were not standardized across patients, reflecting real-world heterogeneity. The therapeutic strategy was selected on an individualized basis, which limits direct comparison between the different agents administered. Moreover, unlike in clinical trials, the follow-up protocol was influenced by real-world practice and by the diagnostic methods available: In locally advanced colon cancer the timing of response assessment was determined by the initial surgical objective, whereas in LARC follow-up followed the response-assessment schemes established for neoadjuvant strategies. This limitation is not exclusive to our series but is shared by the current literature, where no consensus exists regarding the optimal ICI regimen, treatment duration, or surveillance strategy. Additionally, MMR results were unavailable at ICI initiation in two LARC patients, limiting analysis of their optimal treatment trajectory. Furthermore, dMMR status was determined exclusively by immunohistochemistry, which, although widely accepted in clinical practice, was not confirmed by orthogonal molecular methods such as MSI-PCR or next-generation sequencing. Notably, patients 3 and 4 harbored a BRAF V600E mutation concurrent with MLH1/PMS2 Loss, a pattern highly suggestive of sporadic dMMR secondary to MLH1 promoter hypermethylation rather than lynch syndrome[24]. However, methylation-specific testing was not performed in these cases, representing an additional limitation that should be considered when interpreting the mechanistic basis of dMMR in this subgroup. Despite these limitations, real-world series are indispensable for documenting feasibility, characterizing response patterns in unselected populations, and identifying practice gaps that clinical trials cannot address. Above all, this series reflects a genuine real-world experience that complements controlled trial data by showing how immunotherapy-based strategies perform when applied in routine surgical oncology practice.

The growing body of evidence supports the integration of ICI-based strategies into routine multidisciplinary care for patients with non-metastatic dMMR intestinal cancers. Our experience further emphasizes that systematic molecular characterization at diagnosis is a prerequisite for identifying eligible patients, maximizing the clinical benefit of these therapies, and facilitating individualized surgical management. As shared decision-making becomes increasingly central to cancer care, the prompt translation of high-quality evidence into routine clinical practice is essential to ensure that patients can access the most effective treatment strategies available.

CONCLUSION

Immunotherapy is reshaping the management of non-metastatic dMMR colorectal and small bowel cancers, with benefits that extend beyond clinical trials into routine practice. Our findings support the feasibility of individualized treatment strategies in real-world settings. Systematic MMR testing at diagnosis, combined with expert multidisciplinary decision-making, is essential to ensure timely access to the most effective evidence-based treatment.